Abstract:Cyanobacterial water pollution has been threatening the cleaner ecosystem and urban sustainability due to the harmfulness to aquatic ecosystems and human health, which triggers the development of an effective forecasting tool for cyanobacterial blooms. Along the water column, the variations in cyano-bacteria cell densities show various distribution patterns and are influenced by multiple environmental factors. Most data-driven models treat cyanobacteria forecasting at a specific water depth as a single task, which fails to share knowledge amongst water depths, resulting in unfavourable forecasting accuracy. This is why an increasing number of nonlinear black-box models have been built for cyanobacteria forecasting but at the expense of model interpretability. This study aims to investigate whether fore-casting accuracy and model interpretability can be enhanced by (i) using easily accessible predictors and (ii) developing a feature reconstruction-based multi-task regression model with knowledge sharing amongst water depths. Real-world data from a tropical lake are used to evaluate the effectiveness of the model. For the studied lake, the highest average cyanobacteria cell density occurs at 1.0 m, after which it decreases by over 30% at 5.5 m. The correlation coefficients of time-serial cyanobacteria cell densities between adjacent water depths are greater than 0.95 (P < 0.001). The forecasting results indicate that, compared to single-task nonlinear models, 20.59%, 16.25%, and 22.70% error reductions, measured by the mean square error, are achieved for one-day-ahead, two-day-ahead, and three-day-ahead cyanobacterial bloom forecasts. The accurate bloom and non-bloom signals under the proposed model are up to 94.81% and 98.28%. Based on the proposed model, the relative importance of predictors, the sparsity of regression coefficients, and the covariance relationship of regression coefficients can interpret the model adequately and elucidate the mechanism of knowledge sharing and forecasting accuracy improvement. (c) 2021 Elsevier Ltd. All rights reserved.

A Bayesian Structural Model for Predicting Algal Blooms

Probabilistic Prediction of Algal Blooms from Basic Water Quality Parameters by Bayesian Scale-Mixture of Skew-Normal Model

Predictriver Water Quality Based on Bayesian Hierarchical Model

Algal community structure prediction by machine learning

Harmful Algal Blooms Prediction with Machine Learning Models in Tolo Harbour

Deep-Learning-Based Approach for Prediction of Algal Blooms

Algal bloom forecasting with time-frequency analysis: A hybrid deep learning approach.

Modeling of algal blooms: Advances, applications and prospects

Causality Analysis and Prediction of Riverine Algal Blooms by Combining Empirical Dynamic Modeling and Machine Learning Techniques

Predicting coastal algal blooms with environmental factors by machine learning methods

Red Tide Time Series Forecasting by Combining ARIMA and Deep Belief Network

An approach of improved dynamic deep belief nets modeling for algae bloom prediction

A Feature Reconstruction-Based Multi-Task Regression Model for Cyanobacterial Distribution Forecasting along the Water Column

A Novel Single-Parameter Approach for Forecasting Algal Blooms.

Prediction of algal bloom occurrence based on the naive Bayesian model considering satellite image pixel differences

A Framework Based on Hidden Markov Model with Adaptive Weighting for Microcystin Forecasting and Early-Warning

Vertical Development of a Prorocentrum Donghaiense Bloom in the Coastal Waters of the East China Sea: Coupled Biophysical Numerical Modeling

Revealing Physiochemical Factors and Zooplankton Influencing Microcystis Bloom Toxicity in a Large-Shallow Lake Using Bayesian Machine Learning

Application of Improved Core Vector Machine in the Prediction of Algal Blooms in Tolo Harbour

Unsteady Multi-Element Time Series Analysis and Prediction Based on Spatial-Temporal Attention and Error Forecast Fusion

Multi-factor Identification and Modelling Analyses for Managing Large River Algal Blooms.